Please also see Egg Study Redux: Correcting the Stats
Dr. J. David Spence of Ontario’s Stroke Prevention and Atherosclerosis Research Center and two of his colleagues have just published a study in the journal Atherosclerosis purportedly showing that eating egg yolks is almost as bad for our arteries as smoking cigarettes.
This paper is more like an aggregation of clinical notes than what we would expect from something we’d call a “study.” In 1995, they began measuring total plaque in the carotid arteries of patients who were referred to their vascular prevention clinics. Prior to the year 2000, they gave the patients a particular lifestyle questionnaire that they tell us very little about. In 2000, their clientele changed. After this point, patients came to them on an “urgent” basis because they had just suffered a stroke or a transient ischemic attack. They gave these patients “a more limited set of lifestyle questions,” which they also tell us very little about.
They asked these patients not only to estimate how often they eat eggs, but to remember for how many years they had been eating eggs at a similar frequency over their lifetime. This allowed them to calculate the number of “egg-yolk years” each patient had accumulated. We should find it unsurprising, then, that patients who had accumulated the most “egg-yolk years” not only ate more eggs, but were also older. Here’s a graph I made from the data in Table 2 showing the astonishing fact that people who had been eating more eggs for a longer time had also been alive for a longer time:
Dr. Spence and his colleagues offer us no validation of their questionnaire’s ability to estimate lifelong egg consumption, nor of the ability of recent victims of stroke and transient ischemic attack to accurately recall their dietary habits. It nevertheless seems likely to me that people over the age of 65 had indeed been eating eggs for longer than people under the age of 55, so the questionnaire is unlikely to be entirely without merit.
It may seem obvious that we have a problem here. Arterial plaque accumulates with age. Couldn’t an association between “egg-yolk years” and arterial plaque simply reflect this accumulation with age? The authors solved this little problem by making a statistical adjustment for age. After correcting for age, a greater accumulation of egg-yolk years was associated with a greater amount of plaque.
This raises the question of why they didn’t just take the simpler approach of comparing those who reported eating eggs more frequently to those who reported eating eggs less frequently. Surely the patients could more accurately recall their recent intake of eggs than their lifetime consumption of eggs, particularly if they had simply been asked something like, “how long have you been eating eggs?” The reason for using “egg-yolk years” adjusted for age instead of “eggs per week” can be found in this statement:
The total plaque area among people who consumed 2 or fewer eggs per week (n = 388) was 125 +/- 129.62 mm2, whereas it was 132.26 +/- 142.48 mm2 in those consuming 3 or more eggs per week (n = 603).
If we put this into graphic form and do our own statistical test, we can see that the difference is clearly small and not statistically significant:
The most disappointing thing about this collection of clinical notes, however, is that time itself is never considered a confounder. Are we really supposed to believe that the switch in the year 2000 to “urgent” clientele who had just suffered from transient ischemic attack or stroke makes no difference? Is it not possible that these “urgent” cases tended to have more plaque? And is it not possible that egg consumption was higher after the year 2000 than before? Here’s a graph of changes in Canadian egg consumption over time from this paper, with a few markings I have added:
The changes in this graph are exaggerated because the vertical axis hits a bottom at fourteen rather than zero. Nevertheless, it would appear that Dr. Spence and his colleagues began collecting their clinical notes in 1995 when Canadian egg consumption, estimated from food disappearance data, had reached a bottom. After this point, the egg’s reputation began recovering and consumption increased throughout the time period in which these clinicians collected their data. The “urgent” cases populating this data after 2000 may well have been eating a few more eggs than the less urgent cases that populated the data from 1995-2000. They may have been less likely to recall their diet accurately having just recently suffered from a transient ischemic attack or stroke, and they may have had a tendency to exaggerate the magnitude or duration of their egg consumption if they thought eggs may have been to blame for their medical condition.
I’m going to continue eating eggs, but if this study has made an egg-free convert out of any of you, please let me know why in the comments!
Please also see Egg Study Redux: Correcting the Stats
Read more about the author, Chris Masterjohn, PhD, here.
My n=1 says not in my case. I have eaten an average of ten eggs a week most of my life, and that’s whole eggs eaten in a meal, not including eggs in prepared foods. I am now 62 and do not have heart disease. THEREFORE (lol), eggs are not a factor in causing heart disease. (And I’m not superstitious about saying so. ;o) Either that, or whatever else I’m doing is working to prevent the insidious effect of all those eggs!
Oh, and by the way, I’ll never forget, when I had a stress echo done at a physical, age 54, I happened to tell the technician that I regularly ate lots of eggs and saturated fat. She was in her late 20s and nearly had a stroke when I told her (ok, not literally, lol). I passed the test with flying colors.
Is/was it hard to have conversations with your dr? Assuming he was more traditional western trained.
My doctor is a mixed bag. She knows what is bogus, but still has a hard time giving it up. And the statin salespeople have done a great job on her.
I’ve had some interesting conversations with my (fantastic) family doc. His name is Dr.Spence, son of the Dr.Spence in question. I’ve met Dr.Spence Sr. and I can assure you all that he has neither horns nor cloven hooves, nor was he sporting a whole lot of pharma schwag when we met. He does, however, seem to have a view that everyone is a cardiovascular event waiting to happen and is completely resistant to any lifestyle change (not bad assumptions for a clinician). It’s hard to blame a man for having conviction and a gift for marketing! That said, what kind of egg conversations do I have with Dr.Spence Jr.? I don’t. As long as my A1C, CRP, BMI/waist ratio, and overall affect don’t move in the wrong direction we’ve got nothing to talk about except for overuse/sports injuries, my kids, and the future inevitability of a digital prostate exam! We all have to manage our own health.
I think the only way they can really tell if you have heart disease is by doing a heart cath which is typically not worth the risk. I have two friends age 55-62 that have passed the stress test and echo and had low cholesterol levels and one was an Ironman Triathlete. Both of these guys had a heart attack anyway. So are the eggs and meat and oil really worth it? To me no.
This study scared the hell out of me.. I’m not eating enough eggs!
darn tootin – i don’t eat enough – joking aside, eggs are a staple in our household and this study just adds to the pseudo-scientific stench that radiates off almost every epidemiological study out there –
i mean really – have you EVER seen one that was worth the paper it was written on?
I thought the eggs might be protecting them from heart disease. Since all the participants after 2000 had suffered a stroke this does not compare healthy people to people with heart disease. since the group that ate the most eggs were on average 69 years old when they had a stroke and the group that ate the least were 55 when they had a stroke it could be said that higher egg consumption might be able to delay having a stroke by 14 years.
Hi Glen,
That’s a possible interpretation, but it seems less likely given that egg-years was associated with increased risk after adjusting for age.
Chris
I have to agree with Glen. While it may be the case that risk factors are increased, it looks like one can do all sorts of risky things, get a bunch of plaque, and then STILL not get the predicted stroke until waaay later, by eating a few more protective eggs. Then again, gosh, it sure would be nice if we could see what was going on with all those fine people who never ever had to visit the center.
Regarding first table , I don’t understand the correlation between age and eggs per week which is the right side of the chart. Why would an older person necessarily eat more eggs per week? Am I missing something?
Hi Brian,
The first figure does not show a correlation with anything. It shows that as we move from left to right from lower to higher quintiles of “egg-yolk years,” both age and eggs/week increase in similar proportions. The meaning of this is given in the title of the figure.
Chris
Hey Chris, Would it be fair to say the low egg group was probably more inflamed since they had a stroke with much lower levels of plaque than the high egg group?
It’s theoretically possible, but there is no way to know whether it is true that those who had a stroke had much lower levels of plaque in the low-egg group because they don’t tell us who did and who did not have a stroke.
Chris
Gosh, using egg yolk years without adjusting for age. With that kind of measure you could conclude anything. Vegetable years is will probably be associated with atherosclerosis too.
I can see the headlines ‘water causes heart disease’, ‘breathing causes heart disease’, etc
Hi Steven,
They did adjust for age.
Chris
yes – but did they adjust for a lifetime of eating other foods/substances (and list them with all the assumed accuracy of the egg consumption memories) – foods or health events that damage the arterial system? for example did they factor for modern wheat consumption considering how it has been called into serious question with Denise Mingers reassessment of wheat consumption increases from CHina Study Data?
thanks for taking the time Chris – can’t be very satisfying to destroy such an easy target…
Hi Ravi,
No, they published absolutely nothing about other foods. You are right, this was very boring.
Chris
Just a note on how the rest of one’s diet can affect health. My ex-husband has eaten a very low fat, low meat, high sugar/refined foods diet for years (Which used to drive me nuts!). At 43 he had a stroke and is trying to heal significant heart disease. Maybe if he had a few eggs it could have helped him!
Dr. J. David Spence and Alfred Hitchcock have the same phobia!
was it not the eggs but the birds that freaked Hitchcock?–
And just what sort of eggs are these people eating? There’s a motley raft of hens out in the yard laying a stack of eggs that need eating and they are all a bloody hodgepdge of big and little, brown and white, hidden under bushes for days… Maybe eggs are eggs, I dunno. I like em all.
Hi Kem,
Yes part of the problem is we have no idea where the eggs came from and what they were eaten with. I think that is a relatively minor problem, though, compared to the major problems with the study design and data analysis.
Chris
I’ve had a very low egg-consumption during my first 54 years. Then, at age 54, I had a TIA followed by a stroke. I’ve also had my carotid plaques surgically removed. The right carotid whas 75% plugged. It was also discovered that I had diabetes type 2.
Being a bit of a contrarian, I began eating lots of eggs and saturated fats and have never felt better. Since my plaques are now gone, it’ll be interesting to see if I can get another 54 years, but this time on a very high egg-consumption 🙂
Hi Ulf,
I’m glad you’re feeling better! Thanks for sharing!
Chris
Funny how it wasn’t made into big headlines how egg eaters had among the lowest cholesterol, higher HDL and lower triglycerides…. Guess that wouldn’t push their agenda though.
Is it possible that egg consumption at a time when eggs were black listed could actually be a surrogate marker for unhealthy behaviour generally? Perhaps the egg eaters were fond of junk food as well, and ate a very high carbohydrate/sugar diet along with their eggs.
Hi Maria,
It could be, but as I mentioned, the reputation of the egg was on an upswing during this time. Nevertheless, it was on an upswing after reaching a bottom in the 1990s, so it is tough to say. They provided no information about any other aspects of diet or behavior.
Chris
This sounds the most plausible to me. Just because those that follow Weston Price are more likely to eat lots of egg yolks and eat healthy doesn’t mean the rest of the population eating eggs also eat healthy. There is a large segment of society that no longer knows what to eat! They’re tired of the conflicting messages. Or they just never cared to begin with. These people eat lots of sugar, trans fats, and rancid vegetable oils, with their morning eggs and cigarettes. Eating a lot of eggs would never make up for a diet like this and the person on a low fat, moderate sugar, no egg diet would probably fair better.
The study is seriously flawed and has nothing to do with a low carb / high (good) fat diet.
You brought up a lot of the issues that I saw with the “study” as well! I missed that they corrected for age when reading the paper, but I guess they did. I still don’t think it is a valid “design” though. I think the main issue is the measurement of “egg-years”. I can see how this is a way to correct for the length of time at that egg consumption level, but age is inevitably a factor in athersclerosis and I think egg-years is more a measure of age then anything. I think this report is also flawed with how little OTHER information we know about these people. Acitvity level, what made up the rest of their diet, etc. Were they eating white wheat toast with those eggs?
This study contributes to nutritional sciences on some level, but did not deserve the media frenzy that resulted. The news reports made it seem like a randomized, double blind, placebo controled clinical trial had found that eggs were literally as harmful as cigarettes. That is just not the case…not in the least.
Hi Danielle,
I don’t think this study contributes to nutritional science at all, but I think it would be useful to use as preliminary data to apply for a grant to conduct a useful study where the relation between egg intake and total coronary plaque is studied.
I agree with you that the use of “egg-yolk years” is deeply problematic, but I think it is mainly because of the inaccuracy that is likely introduced. It’s relatively easy for someone to answer the question, “when did you start smoking?” And calculating “pack-years” is going to be very imprecise because the person doesn’t necessarily maintain the same smoking habits over time. But someone is even less likely, I think, to maintain the same “egg habit” over decades, and most people don’t have a clear “start date” for egg consumption, so I think trying to calculate this dimension just introduces error. Given that it improved the association, it is probably biased rather than random error.
I agree with you that most information we would want is not included.
I really think the main issue is not the “egg-yolk years” but rather the fact that this is a collection of clinical notes. No one would *design* a study for its own sake that would select participants like this. They are not a random sample of any population or subpopulation, and the selection criteria changed through the course of the study. That introduces all kinds of potential biases.
Chris
Would you say then there is a weekly threshold? As in too many eggs per week? I eat 3 a day.
Robert
Hi Robert,
I suppose there could be but I don’t see a reason to think that.
Chris
the only thing that hangs in the back of my mind is something wonky ol mercola said about being careful no to develop an allergy to eggs by too high a consumption – has anyone experienced this? (cause we too eat maybe 2-3 a day 4 or 5 days a week with no ill effects for years)
Why isn’t p=.42 considered statistically significant? I thought that was always considered significant (I am in a stats class right now, so this is actually helpful stuff :))
Hi Taylor,
Generally something is considered statistically significant when the p value is less than 0.05.
Chris
I know nothing about stats, but are we talking 0.042 vs 0.05 or 0.42 vs 0.05, which are actually the two numbers stated here.
Hi Darcie,
The p value is 0.42, so it is not even remotely close to statistical significance, which would be below 0.05. Does that clarify?
Chris
When will people stop just looking at things like LDL and HDL. The culprit that needs to be studied more is pLDL. The evidence shows that this is the particle that put nicks in our arteries causing the plaques to form. This pLDL seems to be coming from a problem with blood sugar regulation.
On the other hand we keep talking about saturated FAs. They produce “light fluffy” LDL that seems to be harmless to arteries. The whole system needs to readjust how they look at heart disease. That might take cooperation with the food industry……that was a pause for laughter.
Hi Chris, thanks for this post. I just would like to point out an important problem related to this: “After correcting for age, a greater accumulation of egg-yolk years was associated with a greater amount of plaque”. Controlling for confounders is nice, but as you’ve shown through the graph, age and egg-yolk years are very strongly correlated, to the point of being collinear with one another.
Here is the problem: whenever you introduce collinearity (a.k.a. multicollinearity) to a model by adding variables as predictors (or independent variables), you can severely distort the results:
http://bit.ly/NoVr91
http://www.youtube.com/watch?v=avPWO324E0g
The authors should have checked for collinearity, which is not hard to do. If they found that age and egg-yolk years were collinear: (a) they could not use both variables in their model, because they were measuring the same “thing”; and (b) they could not safely conclude that the “thing” they were measuring was egg-yolk consumption, as it could be age.
What they could have done, as you noted, was to control for egg-yolk consumption frequency; or use it as their main predictor with age as control. As egg-yolk consumption frequency had an insignificant effect, however, the authors would be left with the trivial conclusion that aging leads to plague buildup. The paper would then likely be rejected.
Just wanted to leave a quick note to tell you how much I enjoyed reading that!
This is a correlative study – thus it could be that the association is something else that is eaten at the same time – toast with jam(high carbs),bacon – toast etc. Or other habits of egg eaters ( they smoke more ). People were told for years to avoid eggs for good health – I think it is likely that the same people that avoided eggs were also avoiding other things and exercising, quitting smoking etc..
These were also patients that had suffered TSAs – I would expect this group to have more people with e3/e3 gene types – and they would have trouble no matter what the source of saturated fats came from.
The conclusion of the paper shows that the authors do not understand what science is – correlations do not show cause and effect. An example of cargo cult science.
I doubt that eggs are bad as we have had a lot of time to evolve to eating them – eggs contain all 8 essential proteins and there is Palo evidence that we were stealing birds eggs for a long time. It could be that the eggs we have now are different – they feed lots of corn to egg laying hens – and that produces lots of O-6 in the eggs.
In the mean time, we don’t know what this correlation means. That would take a controlled experiment – which a survey is not.
I have a hunch that it may have to due something with the industrialization of our food supply – much of feed comes from dent-corn which is a man made food. The other main sources of calories come from soy and wheat – all three of these sources have been bred and engineered into food that is quite different from what we evolved to eat.
Hi Karl,
Yes, at most the study could tell us there is a correlation. However, this study does not even do that.
Chris
Two the researchers have extensive ties to the statin industry!! Here’s a good review of the study from Mark Sisson:
Hi Dave,
I’m aware of that, but I didn’t mention it in part because I don’t see how that could motivate them to bash eggs. In other words, there is no incentive for the statin industry to decrease the intake of eggs, when that is likely to have little if any effect on statin sales. Since eggs increase total cholesterol in about 1/3 of the population while having little to no effect on others, and since total cholesterol is often used as a criterion for statin administration, then the campaign against eggs is more likely to decrease than to increase statin sales if it has an effect. Thanks for passing on the link!
Chris
Actually, statin producers would want people to eat *more* eggs if they really believed eggs cause dyslipidemia. If that sounds cynical, so be it. It seems to me a large part of pharma is to have people behave however they like because if someone lives in an unhealthy manner, they have a drug to sell you for it!
That was along the lines of what I was thinking. People are suddenly scared of another food and run to their doctors who just prescribe pills for it. 🙁
Eat all the eggs you want, and you can always “cover” them with statins.
except that in this study, eating more eggs actually correlated with the best cholesterol & trig levels–so that making people believe that eggs are terrible could ultimately be good for the statin business. so their ties to the statin industry may actually be relevant if… the study wasn’t a steaming pile of science, of course.
Bonita is correct…hahah
Statin makers would want people to eat LESS eggs cos then people will have worse lipid profiles (and hence need statins)
If I remember correctly reading somewhere, eggs have lecithin in them which is a fat emulsifier and needed for brain function…Too many may be harmful, but some each week seemed to be necessary & healthy! Where you get your eggs is More important~fed “franken grains” & cage-bound is NOT healthy! Free range, bug/dirt/grass picking has more nutrients from what I’ve read and believe. What God put on earth is what our bodies are meant to assimilate! Just saying……
Of course, there are bigger problems with “studies” such as these. That is, all eggs are not created equally. I would not want to eat eggs… if they were from conventional GMO omega 6 inflammatory corn and GMO soy fed hens. Even if the study showed that eggs do cause arterial plaque, “eggs” does not tell me much. A pastured organic egg full of natural carotonoids, and more would probably be very different biochemically.
I agree with you, but I think having meaningless data is a worse problem than having imprecise data. If this were a convincing demonstration that “egg yolks” are tied to plaque, then we might ask questions demanding more precision, such as “which egg yolks,” or “is this relation causal?” But you are right, “eggs” is not very specific.
Chris
Hi Chris,
Did you happen to see this article ?
Eggs and Atherosclerosis
http://www.sciencebasedmedicine.org/index.php/eggs-and-atherosclerosis/
justin
Interesting, Justin, thanks. While the lack of increase in blood lipids leaves a big question mark as to the putative mechanism, I don’t think it’s necessarily as damning as he suggests, because these lipids did not predict plaque accumulation either. So, on the whole, there isn’t a very good explanation of any of the material, but it would be worse if blood lipids were a major predictor of plaque and eggs were a major predictor of plaque but not blood lipids. I suppose having no explanation of anything isn’t all that much a better position, but I think the heaviest criticism of the study is the one I offered here, which is that there really isn’t anything to explain.
Chris
I haven’t read all the comments but I think that you’ve completely miss-interpreted this. The first graph shows the effect of egg yolks on longevity. Egg Yolks cause longevity. QED. Next question.
With all due respect, Dr. Feinman, the graph can’t possibly show any such thing because the x axis is quintiles of “egg-yolk years,” not frequency of egg intake. What the graph shows is exactly what I wrote in its upper label: that “egg-yolk years” is a composite measurement affected by how often someone eats eggs and how old they are.
Chris
It was a joke. Probably, not as funny as the original article. If nothing else, as I’ve said before we use quintiles because to plot the individual points you’d need a computer and there expensive…what year is this?
On the other hand, if you do plot y, (x) vs. y2 (x), you will get a straight line, no? Are you sure we don’t have the secret of life, here?
Hello Doc, it is 2012, and while we are probably all using computer – type devices to read this, we still haven’t figured out the wrap-around issue either.
The Feinman comment evaded my sarcasm detector too.
But I now have another science/nutrition blog bookmarked.
It’s all good.
As you noted, this can hardly be called a study. Both the methodology and the conclusions are sloppy and poorly defined. At 76 I’ve never felt better, and this is after a quintuple bypass in 2001. After that frightening experience I changed my lifestyle radically, becoming a VLC, high protein, high saturated fat convert. My grafts are clean after 11 years, thus giving lie to the lipophobes — including my former internist.
“The mantra ‘eggs can be part of a healthy diet for healthy people’ has confused the issue. It has been known for a long time that a high cholesterol intake increases the risk of cardiovascular events, and egg yolks have a very high cholesterol content. In diabetics, an egg a day increases coronary risk by two to five-fold,” said Spence.
?!?!?
What’s the putative mechanism?
and…
Where’s there solid research to back up these blanket statements?
Sounds like more extrapolation of causation from epidemiology… not good science. Or am I missing something here?
Be well,
Ben
I’ve done a few extra analyses, and it turns out that they suggest that egg consumption is protective. Maybe I’m doing something wrong. I’ll re-check them, and probably post something on this coming Monday.
I haven’t actually read the paper. As they say in the Army, I’m too short to read a paper like that but my comment is based on the first figure that Chris showed above. I said it as a joke, but isn’t that right that if you plot the two variables against each other, they are positively correlated. Please don’t make me read the paper.
Sorry for taking it too seriously. I’m not sure if they’re correlated because they don’t give us individual data. Above, they are not plotted against each other, but rather they are both plotted against egg-yolk years, which is a composite of the two.
Chris
Reading the study takes few minutes. It tells a lot if you don’t bother to read studies you criticize.
Hi all,
I don’t like the statistical analysis the authors used. The multiple regression model is attempting to “control” for age by setting up the proxy predictor variables “egg-yolk-years” and “pack-years”.
I don’t see how the final model controls for age at all. As you point out, age and egg-yolk-years and pack-years are going to be auto-correlated, to the point probably of constituting multi-collinearity.
Am I missing something?
The model should incorporate age specifically and then use egg-intake as like a categorical variable or something. Of course, this would mean you couldn’t have a neat regression model.
It also appears that pack-years and yolk-years are correlated, although less strongly.
I’m not at all convinced that these data are capable of discriminating between yolk-years, pack-years, and age as confounded variables given the lack of experimental design. If there are any non-linearities in causal dynamics between these variables, this kind of partitioning of variance is invalid.
Other possibilities:
~egg yolk quality matters
~genetic sub-populations respond differently
~at-risk populations differ in metabolic consequences of egg yolks
~dietary context matters
Chris
I am not a statistician, or anything like that, but I think your objections to the study are too narrow and that the broader implications of the study, such as pointed out by Ned Kock and Prof. Feinman, need to be seriously considered.
People create or propose results from studies all the time which not only show the opposite of what they intended but expose the inability of the researchers to actually see what they have done and proved. Simply objecting to the study for its numerous confounders and misguided correlation doesn’t help at all.
Secondly, the idea that correlation studies cannot prove anything has got to be, as I see it, the most seriously wrong, and misguided conclusion you present: The idea seems to be that properly conducted tests will “prove” something. The problem is not that correlation studies are so weak, it is that the “gold standard” double blind etc. studies are themselves not really very good, and worse, with the misguided application of the “truth” related terms that are included in the full meaning of “proof.”
“Proof” in this case is similar to “causation,” and “truth,” and about as difficult to get to, in the philosophical context. Any discussion of such terms leads one directly to absolutism and starts running into issues that are mathematical in nature essentially impossible to resolve.
Every study has its weaknesses, and this one has more than most. Even so, it still manages to say a lot. The objection that we should should object because the data are meaningless is wrong. The easy part of the problem is that these clinicians had information and didn’t know what to make of it.
Second, any scrap of information (correlation) can properly lead us to a conclusion, and it may be that all we will ever have is correlation information. It is not a matter of correlation being meaningless, or only suitable for proposing hypotheses. The problem is more profound, because there is no (agreed definition for) proof, just as there is no (agreed definition for) truth. If that is true, as it clearly is (insert smiley face), then we can certainly consider a limited truth (as they all are) based on a series of correlations.
Otherwise this turns in a grail seeking exercise.
Hi Richard,
I don’t think I’ve ever suggested that we should look for “proof” of anything in empirical science, nor have I ever argued that correlations are meaningless. They are not. I’m interested in Ned’s analysis. Dr. Feinman is, I think, misinterpreting the figure I made, not because it is “meaningless,” but because the meaning he attributed to it is incorrect. The meaning of the graph is printed as its title.
Chris
Again, my comment was meant as a joke although if you do know egg consumption and age separately, you may be able to use my interpretation. To be very serious, what is wrong with this study is the reporting in the press. Why have we paid for expensive large studies (and small studies) that consistently failed to show any effect of dietary cholesterol or dietary saturated fat. What is the purpose of the scientific literature if you can ignore it at your pleasure? Whatever the investigators found, they should interpret it in terms of the near complete failure of the diet-heart hypothesis. in other words, the result, if it meant anything at all, should be described as a surprising anomaly, at odds with the bulk of the literature. So, I am certainly in agreement with Chris and everybody on the basic idea. So my joke wasn’t that funny.
One thing conspicuously absent from this new paper is a Conflict of Interest statement.
Here’s the one from their 2010 anti-egg paper:
Dietary cholesterol and egg yolks: Not for patients at risk of vascular disease
Can J Cardiol 2010
Conflict of Interest
Dr. Spence and Dr. Davignon have received honoraria and speaker’s fees from several pharmaceutical companies manufacturing lipid-lowering drugs, and Dr. Davignon has received support from Pfizer Canada for an annual atherosclerosis symposium: his reasearch has been funded in part by Pfizer Canada, AstraZeneca Canada Inc. and Merck Frosst Canada Ltd.
Pfizer (Lipitor), AstraZeneca (Crestor), and Merck (Zocor), have much to gain if people are terrified of cholesterol. These types of papers serve their interests well.
BTW, they again invoke the name of Ancel Keys in this new paper, but they’ve forgotten how to spell it.
Now he’s Ansel[sic] Keys. :rolleyes:
Be well,
Ben
You got it! If they can show “correlation” between some bad health outcome and eggs, then they can somehow slide into the idea that it is from the cholesterol, then they can slide into the idea that cholesterol-lowering drugs are of some benefit, despite other data showing otherwise. I am also looking forward to Ned’s analysis and Feinman’s sense of humor.
I’m eating a lot of eggs now, to compensate for the effects of the statins.
Exactly. The entire purpose of this silly “study” is NOT to prove that eggs are dangerous.
It’s to reinforce our fear of cholesterol, and thus increase our willingness to take statins.
– end of story.
I am not a scientist nor do I have a PhD in nutrition. However, just from a layman’s perspective, did the study have a group who self reported eating less eggs under the exact same conditions?
Also, the relationship of the authors to the pharmaceutical companies should have been disclosed. I thought that they havea to disclose that conflict according to certain professional guidelines?
Finally, here is my dumb non-scientific question. Why is it that after decades and decades of research by experts from every discipline showing no relationship between dietary cholesterol and blood serum cholesterol, that we cannot move on? Am I missing something?
Another great blog post. Whenever I read you ripping int these “studies”, pointing out the most shockingly blatant gaping holes, I can’t help but wonder about the original publication and the processes behind it. Atherosclerosis is a peer reviewed scientific journal, is it not? Where exactly are the peers in all of this?
I think what Scott says is remarkable. Were the peer reviewers in sleep?
Peer review as practiced is often one of the greatest impediments to the progress of science. Most reviewers are inclined to rubber stamp approval of anything that matches their own biases–especialy if it happens to cite one of their own papers.
I say this as someone who has been on both sides of the fence (reviewed and reviewing). The system is broken, and it gets worse the further you get from the hard sciences.
Ever hear of this study?
Fernandez ML. Dietary cholesterol provided by eggs and plasma lipoproteins in healthy populations. Curr Opin Clin Nutr Metab Care. 2006;9:8-12.
“In over two thirds of the population, then, egg consumption leads to little or no change in cholesterol at all. In less than a third of the population, total cholesterol goes up, but both the ratio of LDL to HDL and the total number of LDL particles remains the same; the LDL particles just get bigger and safer.”
This makes me concerned that eating eggs may be bad for the eyesight. I wonder if non-egg-eaters can read the numbers in the third chart.
I eat 3 to 5 eggs a day but my biggest concern, since I am on a limited budget, is that I am eating commercially-raised eggs. Since those chickens are probably fed soy and corn, the eggs I eat are probably high in omega-6 fatty acids and could lead to more LDL oxidation and arterial plaque. I love eggs and would hate to give them up.
There are quite lots of folks who always use improper eyes to look at the fashion. In fact, there are quite lots of women who step into the wrong place when they want to catch up with the fashion trend. They do not know that both their feet and the knees are suffering a great stress while they are wearing a pair of high heeled shoes, in other words it means that they are just losing their health day by day.
Not sure if mentioned above, but the egg yolk as we know if the most nutrient dense aspects of one of nature’s healthiest foods! The yolk has choline a nutrient that is anti-inflammatory and reduces homocysteine levels; cardio protective, aids in cognitive function, and nerve health!
By the way, the authors measured eggs. Was this converted to egg yolks by the “you are what you eat” principle?
Most people I know who care about their heart healthy have been limiting egg yolk consumption for years. In fact my father was told by a doctor to limit consumption way back in the late 1940’s! That factor alone could easily account for the correlation found in this “study”. Of course far too many media outlets have reported these findings without any evalutation.
The other holes in the study are, as Proff Feinman says, that the eggs are somehow being magically separated and there is no effect whatsoever of egg white (or of fats used to fry eggs, butter and salt, “soldiers” and so on.)
Egg white is both a common allergen and a source of avidin so testing for a harm from it wouldn’t be completely far-fetched, were the authors not blinded by cholesterol.
And eggs (yolks, yeah right) are not being compared with anything else. A few controls, like alcohol, smoking, vegetables, fruit juice, bead and pastry would have been easy to collect.
They have also designed their own little quirky method, so that their study can never be compared to others or become part of a meta-analysis.
These people are fools, but the media has been the real villain.
…they may have had a tendency to exaggerate the magnitude or duration of their egg consumption if they thought eggs may have been to blame for their medical condition.
I think that would be an important effect. I’d like to see the survey to see what other foods were asked about.
Consider you’ve had a stroke, come to this clinic, and are presented with a survey inquiring about egg intake. You’d naturally seize on that as a cause of your problems.
After all, you’ve been told for years that eggs were bad for you. And here is this trusted clinic asking specifically about them. You might be tempted even to ascribe to the egg the years you smoked and don’t want to admit to.
Yes, unfortunately they did not publish the survey.
Chris
I suspect these doctors believe truly (as do most cardiologists) that large egg consumption causes heart disease. I suspect they looked at ways from their patients data to “prove” this.
To really gain some excellent data, a longitudinal study with large patient numbers would be needed (thousands of people, across multiple socioeconomic levels ideally across multiple countries over greater than 10 years). Then variables of baseline health, fitness, consumption (of everything, organic or industrialized) etc need taken then measured. Can’t see that being funded by anyone! Not even the egg industry who may have something to gain!! A nice dream though as I think this would yield some fascinating information that really would provide some worthwhile data.
Hi Michelle,
I think you are right on the motivations, but large prospective studies have been done with eggs, and small randomized trials have been done with eggs. This “study” is an anomaly in terms of the results, but it is also an anomaly in terms of being incredibly poorly designed. I think the relation between the two is more than coincidence. 🙂
Chris
“I’m going to continue eating eggs”
This makes me sad. If you have many followers, I won’t see any drop in prices of pastured eggs.
I eat 3-4 eggs for breakfast every day and will continue to do so there are many nutrients in the yolk of the egg. I also believe cholesterol is good for brain health
Hi Chris,
So, not looking to rehash everything here about the study having little to no contribution to science, but wondered about three quick things.
1) Am I misreading the results in the abstract? Aside from everything else, is the margin of error on the TPA measurement really greater than 100%?
2) They did adjust for age in the egg yolk years, but how does that work out mathematically? Is that based on TPA on this population, or what’s been done elsewhere?
3) I saw a comment above and elsewhere that cholesterol actually improved with egg-yolk years — I didn’t see any blood analysis in the abstract, and not paying $30 to see the full study, but was that mentioned in there?
Thanks in advance
Jason
Hi Jason,
That’s the standard deviation, not the margin of error. SDs always look big. The error bars in the graph I made show the standard errors, which is a measurement of statistical precision that incorporates both the SD and the sample size. As you can see, they are not very large.
As far as I can tell, they used the correlation between age and plaque internal to the study to make the adjustment.
Changes in cholesterol between quintiles were marginal and not linear. If you go to Zoe Harcombe’s blog she reprinted the table.
Hope that helps,
Chris
thanks. the age adjustment would seem to be even more flawed in that case, right?
guess i’ll have to revisit my statistics from college. seems rather large though.
thanks for the link to zoe’s blog and study data. pretty enlightening to read the actual data.
saw your comment about self-interest with the researchers. i agree, it doesn’t make logical sense, but there has to be something there. #1, most people aren’t logical, only some of us are most of the time. #2) i find it hard to believe something this shoddy get’s without self-interest of some kind, even if it’s a backwards logic.
Hi Jason,
I don’t think the age adjustment is flawed per se, I think the fact that they introduced it as a confounder in the first place is the problem. The association is driven by the answer to the question “how many years have you eaten eggs for?” and not “how often do you eat eggs?” How do people come up with a sensible answer to the first question? Most people have been eating eggs as long as they can remember, no?
I think often times the “logic” of self-interest goes something like this: 1) I believe cholesterol is evil, 2) I’m going to help people with that knowledge, 3) I can get paid doing that.
Chris
Thanks for clarifying this study a bit for me. It confirmed a lot of my own doubts about Spence’s paper here.
What happened to the rigor of the phrase “peer-reviewed”? It doesn’t seem to mean much anymore if a shoddy paper like this gets through and makes national headlines.
BTW, they need to fix the comment threads on your blog — the replies are all squished together and hard to read.
Hi Zachary,
I’ve been corresponding with the guy who runs the site and he’s going to try to get rid of the right margin on specific blog posts. Hopefully that will make enough room to make threaded comments readable.
Chris
Hi Jason,
I’m going to have to respectfully disagree slightly with Chris M’s take on the validity of the yolk-years measure. I don’t think it adequately controls for age, even if we believe the recall of the participants is sound. What it does mathematically is collapse the variability attributable to age and to egg-frequency into one predictor variable for the purposes of regression. Ditto for pack-years, and smoking + age.
The crux of the argument the researchers then present is that, since the qualitative shape of the response curve (arteriosclerosis) to yolk-years appears to be exponential rather than linear as it is for the age versus arteriosclerosis, this indicates that egg-yolk consumption has some cumulative detrimental action on arteriosclerosis.
But to properly “control” for confounders one ideally separates them into different variables rather than collapsing them together.
I’d like to see a statistical model which explicitly controls for age to understand if there is, indeed, any “signal” at all in these data. Given the large error bars in the regression of carotid thickness on yolk-years, I suspect the relationship may actually be linear, and entirely due to the confounding effect of age, and possibly a small colinearity with smoking (i.e. people eating more eggs maybe more likely to smoke).
Chris
Hi Chris,
I think the error bars are large because they are depicting standard deviation instead of standard error. The graph I made converted the SD they reported to SE and the bars are small.
I’m not sure what you are disagreeing about. I don’t think the participants’ recall is likely to be sound, but the model did explicitly control for age.
My main problem with the yolk-years variable is that, since there is no relation to egg frequency, it would appear the relation to yolk-years is driven entirely by the question “how long have you been eating eggs?” Now, in the context of smoking, I understand what this question means. For example, I can remember that I started smoking when I was 12 going on 13, and quit in the summer of 1997 when I was 15. If someone asks me when I started eating eggs, I am at a loss because, like most Americans, I have been consuming eggs longer than I’ve been forming permanent memories. So what does that question even MEAN?
Chris
Hi Chris,
I guess I’m not convinced that it does explicitly control for age, when the effects of age are distributed into the yolk-years and pack-years variables. I think it makes the confounding harder to tease apart than it already is in an observational study.
Your point about the validity of the question is well-taken. I agree it’s a major flaw.
I should add that the only justification I can see for the yolk-years measure is to explicitly test the hypothesis that egg yolk consumption generates a cumulative damage- rather than only a short-term effect highly responsive to frequency of consumption at a short time-scale.
But the authors do not provide such an hypothesis test, or even parameter estimates. Here’s what they say, “Curve fitting with the cases that had non-zero values for egg yolks and smoking showed that an
exponential fit was better than a linear fit.”
To be totally fair, I think there may be something going on here, beyond statistical artifacts. I’m not at all convinced it’s what it appears to be though…
Chris
This comment is for Chris Wilson, and I’m replying here so it doesn’t get too squished.
Chris,
I think I may have misunderstood what they meant by “adjustment for age in a General Linear Model,” which I suppose is part due to my own sloppiness and part due to the fact that their own reporting of their methods is abysmal.
Are you saying that your interpretation of their “adjustment” was simply multiplying the number of eggs consumed by the age of the person? How is this “adjustment for” anything? Moreover, why do they call this “number of years consumed” in the methods rather than “age,” and then claim on page 4 that “Age is incorporated into pack-years and egg-yolk years”?
I don’t understand how this passed reviewers and editors. I’ve seen plenty of bad inferences in science but this is just downright incoherent reporting of their own design.
Now, on the other hand, how could they possibly mean this with respect to pack-years? Did they really just multiply the person’s age by the number of cigarettes as if the person had been smoking since infancy?
Chris
You know what’s pretty funny — if you look at the graph I made of age and egg intake as a function of yolk-years, you’ll see that the increase in egg intake is linear but the increase in age is something that looks sort of like their exponential curve.
So it would appear that this “exponential” curve might be driven by the fact that age increases linear as a function of age, but exponentially as a function of yolk-years.
Chris
My take on the paper (not having read it!!) is that they did essentially the same thing as plaque = f(age, yolks, age*yolks) without calling it as such. Age shows up as significant, it is relabeled as a correction instead of a factor. The yolk*years term is also significant, indicating a significant interaction term. There would be not much of a problem dropping the main effects in a factorial design, since by design, they are independent. Doing that with a mess of data is another matter.
So what if these folks egg meal consists of 20oz coffee (12 cream/12 sugar), 2 slices white toast (with 4tbsp butter substitute) and a jelly infused donut? Ahh… forget that.. It’s definitely the egg thats making the plaque!
Chris Wilson and others,
I emailed Spence, and he told me that for the eggs frequency measurement, they adjusted for age by entering it as a covariate in SPSS, entering egg intake as a fixed factor, and entering plaque area as a dependent variable.
I’ll need to ask him for some clarification on this, but he also told me that because yolk-years and pack-years contain a time element, “it was necessary to adjust for age,” so I am tentatively taking this to mean they explicitly adjusted for it in the model rather than “adjusted” for it simply by eyeballing the graphs to see the shape of the curves.
He also said he expects to provide more data in response to letters to the editor. Unfortunately I imagine that will be roughly a year from now, but it sounds like he expects a lot of criticism, and is aware that the paper as published generates an awful lot of questions. It will be interesting to see how that turns out, though no doubt the media will have forgotten of it by that time.
Chris
Interesting follow-up. I’m looking at Table 3 where they report their multiple regression analysis. It says, “Egg-yolk years remained a significant predictor of
carotid plaque area after adjustment for sex, total cholesterol, pack-years of
smoking, systolic blood pressure, diabetes and body mass index. *(Age is incorporated
into pack-years and egg-yolk years)*.” Also, their definition of yolk-years: “For example, a person who consumed 3 eggs per week for 50 years would have a score of 150 egg-yolk years”.
I’m confused about your first paragraph: entering age as a covariate of “egg frequency” or “yolk-years”? If the former, that implies a trend toward higher-frequency in some age groups than others (this could explain the non-linearity in your graph relating age as a function of yolk-years). I’ll have to think about whether that adjustment would allay my concerns…
It sure doesn’t sound like age was entered anywhere in this analysis explicitly, or used to “correct” the yolk-years measure. Perhaps they simply wrote up their methods/results incorrectly?
At any rate, based on Table 3, the effects of age are lumped into the yolk-years and pack-years measures, presumably inflating their importance in explaining carotid plaque progression. Interestingly, the whole model has an R^2 of 0.27, which isn’t very good IMO, even for biology. I wonder what it looks like with age as an explicit factor?
Chris
Hi Chris,
Age was not adjusted for in the multiple regression.
Age was adjusted for in the comparison of 3/week egg intake frequency. It was entered as a covariate in SPSS GLM, with egg intake as the fixed factor and total plaque area as the dependent variable. As far as I can tell, this is ANCOVA.
Dr. Spence indicated to me that age was not adjusted for because “it is incorporated into” yolk years. I think what this means is that they would be strongly related to one another, and thus one can’t be used to “control” for the other.
Chris
Also, like you I’m surprised that this got through the peer-review; their analysis has so many peculiarities that evidently aren’t particularly well-explained or justified in the paper. It almost seems duplicitous- but I’m certain the authors are genuinely concerned about egg yolks for whatever reasons and are simply trying to make sense of some clinical data.
I would say this paper generates hypotheses that need much better testing, rather than answering anything. Predictably, the media are incapable of portraying nuance like that 😉
Chris,
On a side-note, I’m wondering what your take is on Dr. Attia’s stance on cholesterol and plaque formation. In essence, he seems to be saying that LDLp is what drives the process, and predicts an individual’s risk of plaque progression/development the best. The argument is by appeal to integration under a curve of LDLp, all of which are susceptible to infiltration/oxidation to some degree.
However, what about Krauss’ and others work showing different infiltration risk for different classes of LDL? It would seem that Dr. Attia’s position requires all LDL classes to behave more or less the same in terms of infiltration/oxidation risk (which seems suspiciously like the conventional Lipid Hypothesis). The process is more or less stochastic- a sort of mass-effect that is a function of particle numbers passing over vulnerable portions of arterial endothelium.
Also, you’ve noted before that statistically-speaking, at least, correction for TC:HDL ratios eliminates the predictive power of LDL particle sizes.
Since LDL-c=LDL-p*avg(particle mass)
Shouldn’t the TC:HDL ratio behave similarly wrt LDL-p?
So I’m still wondering if NMR is worth it at this point.
Aren’t there some circumstances when more LDL particles is a good thing? All other things being equal, wouldn’t it be better to have more “large, fluffy” LDL particles, than fewer “small, dense” LDL particles?
Chris
Chris,
NMR is worth the <$100 cost to see what % of you ldl particles are the small dense ones. These are the ones I want to control.
Remember, in your world of n=1, you really won't know for some years what the real story is. I have done enough research to at lest convince myself to keep the % of small dense ldl particles low. It is fairly easy to do.
Hi Rob,
Sure, if you have money for playing around, then there is nothing wrong with playing around.
Chris
Hi Rob,
I think if you have an extra ~$100 for playing around, then by all means play around with it. At the moment, it’s up in the air whether the NMR results provide a reliable way to gauge risk or the success of a dietary or lifestyle change. That doesn’t mean you shouldn’t measure it if you want to though. We’re all entitled to act on our best guesses.
Chris
Hi Chris,
I don’t think that it’s all about the number of LDL particles, as I’ve said before. There is as much vulnerability in the lipoprotein as there is vulnerability in the endothelium, so its composition matters, and as I’ve stressed before I think the best evidence is for residence time. I don’t think particle size is “ready for prime-time” as one review put it. There is no evidence that it contributes independent information about disease risk and I don’t think it provides independent information about the most important biochemical processes. People can intervene based on poorly researched surrogate markers like this if they choose, but there is much more solid research on the total-to-HDL-C ratio, and torcetrapib was a disaster.
Yes I think it could be better to have more particles if they are larger, but in general they will be inverse, because if the particles are smaller than the same mass of LDL will be represented by more particles, and vice versa.
Chris
So… why are they blaming the yolk? How do they know it wasn’t the whites?
To bad most of the sheep reading the article will believe it to be truth.
The paper says something rather interesting about eggs and diabetes.
‘Analysis of the Health Professionals and Nurses Health studies by Hu and colleagues noted no overall effect of egg consumption on CHD events and mortality after 8 years of follow up, among participants who remained non-diabetic [11]. However, among participants who became diabetic during follow-up, daily egg consumption doubled CHD mortality [11]. The same pattern was seen in another large population-based study: among participants who became diabetic during follow-up, intake of eggs was associated with a doubling of cardiovascular risk [38]. Extending these findings further, a later assessment of the Physicians Health Study demonstrated a relation of egg consumption to total mortality and confirmed an even stronger relation to mortality in those with diabetes [13], and a Greek study in diabetics showed that daily egg consumption increased coronary risk more than 5-fold [39].’
Diabetes is associated with iron overload, and egg cholesterol should only be a problem if it’s oxidised. Is this the answer?
Hi Jane,
Yes, that’s probably one of the >>>1,000,000 possible explanations for that observation.
Chris
Hi Chris
If you can give me even 9 other explanations, let alone 999,999, I will … do something difficult of your choosing.
Hi Jane,
I wouldn’t ask anything of you, but:
1) In diabetes, loss of insulin signaling depresses thyroid function, which decreases the clearance of LDL-C and increases its concentration in the blood. Spence is right that high LDL-C is the major driver of atherosclerosis, and the high LDL-C that results from consuming more egg yolks (perhaps largely limited to the postprandial state) causes atherosclerosis.
2) In diabetes, loss of insulin signaling depresses thyroid function, which decreases the clearance of LDL-C. I am right and concentration is not very important, but decreased clearance leads to increased residence time and thus a greater degree of oxidation, which contributes to atherosclerosis.
3) In diabetes, loss of insulin signaling leads to depleted glutathione. This leads to loss of protection against oxidative stress, which leads to greater oxidation of LDL membrane PUFAs, which causes atherosclerosis. Eggs increase circulating LDL-C, if even only postprandially, so there is more LDL to be subjected to this process.
4) Same as 3, but the depressed glutathione leads to greater methylglyoxal (MGO) accumulation, which increases MGO-modification of ApoB-100, which causes atherosclerosis.
5) Diabetes is strongly associated with fatty liver disease. The choline in egg yolks thus allows much greater liberation of lipids from liver into plasma, which contributes to atherosclerosis.
6) Loss of insulin signaling in diabetes depresses thyroid function and depresses formation of bile acids, and decreases gut motility, which leads to poorer gut health, which increases the formation of trimethylamine from the choline in egg yolks, which contributes to atherosclerosis according to a recent study I blogged about (and criticized).
7) Diabetics have a greater risk of renal damage and poorly excrete trimethylamine formed from the choline in egg yolks.
8) Eggs cause atherosclerosis in everyone, but only diabetics have sufficient inflammation and oxidative stress to die from it in measurably greater proportions.
9) Most importantly, there could be a multitude of non-causal explanations: only two studies made the comparison, they could be flukes; diabetics are less likely to report egg intake accurately and some spurious association thus arises; egg intake is confounded by other dietary and lifestyle factors, and so on.
Chris
Wow! Exactly nine! A Masterstroke, I think …
” Much of the controversy in this area is about effects of egg yolk consumption on fasting lipids; however the main impact of diet is on the post-prandial state, not on the fasting state [15], [16].”
This quote is from the introduction. If this is really the case, why isn’t it mentioned and speculated in the discussion section? Especially as LDL did not increase with egg-yolk years.
Hi Reijo,
I think because they didn’t measure it. Usually the introduction introduces the general significance of the topic while the discussion discusses the significance of the findings, and there were no postprandial findings. Keep in mind this isn’t really a “study” as we usually think of it, but rather a retrospective statistical analysis of a large grouping of case notes.
Chris
Hi:
We eat lots of really great local fresh eggs. My husband and all his family have terrible ratios and high as the sky cholesterol when untreated. My husband was told he had blockages growing in his 20s. He is now 57. Our daughter was literally giving him emotionally based chest pains. Based on his family history , his father had two quadruple bypasses by 65 , starting at 51, his mother’s aorta filled and calcified and had to be replaced, his brother had a quintuple bypass last year, so when he was having chest pains that came on worse on the treadmill, they scheduled him for a cardiac cath. Since previously he had been told that it was building they were worried that he would not be able to get a stent through and that bypass was a possibility. We were told 2 1/2 hours was about the shortest it would take.
He has been eating a high saturated fat, meat and vegetable diet , with some low carb fruit for years now. Years. He lost a ton of weight and felt well.
In 45 minutes the doctor was out and looking for me. He told me that he could find NOTHING. He said my husband had the arteries and heart of someone under 20 years old and there were absolutely no plaque buildups at all.
When we first started on this diet his total Chol was 460 with a terrible ratio. We did it for a while and went back. His total was down to 165. My triglycerides went from around 1500 to 145. I am not saying that we can’t have a heart attack . I am 62 and feel that stress can cause them but causing spasms. But my husband to me is living proof that build up can be reversed with a natural diet of meat, milk, dairy, seafood, vegetables and a limit amount of fruit as a diet, including eggs and cheese as a staple. Two of our almost vegan friends at this same time had 100% blockages in their main artery and significant blockages in the others. They were saved by stunting, but it was almost beyond stenting when it was caught. So we are all going to die sometime, and I might die of a heart attack, but it won’t be from an egg.
Congratulations Ann, and thanks for sharing your wonderful story!
Chris
I’ve been keeping informed of Dr. David Jenkins dietary recommendations for decades now. If you check into his publications and the St. Michael’s Hospital, Toronto dietary information, you will not ever be asking ‘Where’s the beef?’ Dr. Jenkins et al.espouse a diet that would make any human being shit like a snail. If you’ve ever owned an aquarium, you’ll know what I mean. In the initial ‘experiments’ most subjects were unable to follow the dietary requirements. They advocate living on oat bran, psyillium, tofu, whole grains, pulses, tree nuts and a panoply of ‘high fibre’ foods which allegedly remove ‘bad cholesterol’ from the body. Dr. Jenkins needs to retire. His ship has sailed.
I was reading the study & wondering about the fact that the BMI in the “Baseline Characteristics” chart was so high: above 27. Would this be a confounding factor that paints a picture of unhealthy habits that was ignored – just like as Ravi pointed out above so were the foods these people ate with their eggs over the years, which probably included white or wheat toast, with a schmear of margarine (industrial seed oils & fake trans fats) & some jelly or jam (high fructose corn syrup)?
Hi Lesley,
I think a BMI of 27 is actually under the national average. No doubt prior dietary history affects how eggs or anything else in the present impacts one’s health, though I think that would become more interesting for a study that actually shows an effect of eggs, and I don’t think you could say that for this study.
Chris
If the authors have connections with the statin industry the purpose for doing this study is quite simple and that is to keep the cholesterol/heart disease association alive.
Hi Andy,
That’s certainly possible, but although I don’t watch much TV, I’ve seen a few advertisements for statins and it seems a major schtick for them is that you may not realize that it’s your genes responsible for your high cholesterol rather than your diet, so you may need medication. I’m not sure how well dietary studies like these help them. I think it’s overly simplistic to assume someone’s salary always determines their ideology. Often times people choose how to make money based on their ideology.
Chris
hi
egg white is peat of a protein.
if a person don’t eat the yolk is that okay?
if they continue lacking the other part of the egg protein, what are the effects?
Hi jj,
Most of the important nutrients are in the yolk, so it isn’t as nourishing. Most importantly, egg whites, even after cooking, have a compound that inhibits biotin absorption, an important vitamin. Egg yolks are loaded with biotin, so it balances out if you eat the whole egg. Some people could possibly develop a biotin deficiency from eating just the whites. I’ve written about this here: http://www.cholesterol-and-health.com/Egg_Yolk.html
I eat 4 eggs every morning (barring running out or rare circumstances preventing me). My total cholesterol is 183, which I consider too low. My HDL/LDL ratio is also too low. But these cholesterol aspects are because I also eat too much sugar.
I just LAUGH when I hear doctors and dieticians talking about reducing their blood cholesterol by reducing their fat & cholesterol intake, and that this is supposed to lessen their heart disease risk. They have no business speaking where they have no knowledge!
Probably one of the biggest confounding factors you did not mention is the ‘recall bias’ associated with prognosis. If you are in a doctors clinic and you are recovering from heart disease or stroke, it is very likely your perception of your lifestyle will be skewed towards a ‘negative’ assessment. Most people know something of the diet-heart theory and will likely blame their heart-disease or stroke on those parts of their diet that they are told have caused it. They are more likely to over-estimate the ‘bad’ components of their diet on a questionnaire.
Your analysis of the nurses study showed the similar but opposite effect where well educated healthy people tended to forget their ‘bad’ days.
Ask a ‘reformed’ heart-disease patient how much fatty food they used to eat, and I bet you they can remember eating bacon, eggs crackling and cheese everyday.
I must wonder at the publishing of these spurious studies. I could not imagine that even if patients are recalling their egg-years correctly, that there could be much statistical significance left after removing the age based correlation.
Are you perhaps taking the bait by responding to these weak studies, and giving these psuedo-science trolls oxygen?
Thanks for your input, Gordon. I did mention that: “. . . they may have had a tendency to exaggerate the magnitude or duration of their egg consumption if they thought eggs may have been to blame for their medical condition.” I suppose I might be taking bait here, but I don’t see how writing a post like this really advances their cause in any way, and I’m certainly not a limiting factor in their media attention.
Chris
if you compare 2 eggs vs. 3 eggs a day what kind of results are you expecting? what you should compare is someone who eats (#of eggs)with eats no eggs. that’s when you get nice curves.
the article doesn’t name one reason why it would even be necessary to eat eggs at all.
so why would I?
Hi Eric,
I really don’t mind one way or the other if you eat eggs. I don’t think it’s “necessary” to eat eggs, but they are nutritious. They are a great source of choline, for example.
Chris
I imagine that these people ate other things besides eggs that could have affected their health as well.
We eat 3 eggs a day about 6 days a week, not including the eggs incorporated into other foods. Along with saturated fat from animal products, coconut oil, and monounsaturated from olive oil.
We also use tehina which has been found to lower triglycerides.
It would be nice to see actual studies showing that eating several eggs a day every day for years does not compromise one’s health.